Improvised Cloud Based Venue Recommendation Using Context aware KNN Method

Abstract

Recommendation systems have seen major evolution in the field of information engineering. Many of existing recommendation systems built their copies on collaborative filtering approaches which create them easy to appliance. performance of most of the existing collaborative filtering-based recommendation system suffers because of many challenges, occurrence of some incompatible objectives or decision variables, such as users’ preferences and venue closeness. In this paper, we estimated MobiContext, a hybrid cloud-based Bi-Objective Recommendation Framework (BORF) for mobile social networks. The MobiContext uses multi-objective optimization techniques for generating personalized recommendations. To tackle the problems affecting to cold start and data sparseness, the BORF accomplishes data preprocessing using the Hub Average (HA) inference model. the Weighted Sum Approach (WSA) is employed for scalar optimization and an evolutionary algorithm (NSGA-II) is applied for vector optimization for providing greatest suggestions to the users about a venue. The outcomes of complete experiments on a large-scale real dataset verify the correctness of the planned recommendation framework.